Results

Figure 6.6 presents an SEM image of a working sample. The wire-bonded contact pads which connect to the TMR elements can be seen on both sides. The conducting lines for the positioning structure come from the top and bottom of the image. The alignment cross is right in the center, and two larger test TMR elements are right and left from the alignment cross.

Figure 6.6: SEM image of the completed sample.

The size of the TMR elements on the sample is: 2$\times$2$\mu$m$^2$, 4$\times$4$\mu$m$^2$, 6$\times$6$\mu$m$^2$ and 8$\times$8$\mu$m$^2$. For test purposes, two extra 30$\times$30$\mu$m$^2$ elements were designed. Exemplarily, the measurements of a 2$\times$2$\mu$m$^2$ element are presented here. Figure 6.7 shows the area resistance and TMR ratio of the element. The maximal TMR ration is 21.8% at an area resistance of about 260 to 320M $\Omega\mu{\rm m}^2$. Again, there is the slow ascending slope at zero field. Measuring the I/V curve of the element (see figure 6.8(a)) and applying the BRINKMAN fit to the differentiated plot (b) gives the following results: barrier height of $\nu =$ 1.25eV, a barrier thickness of $b =$ 2.65nm and an asymmetry $d\nu = -0.24$eV. The other elements show similar values for the BRINKMAN fit and the TMR ratio.

Figure 6.7: Major loop of a typical 2$\times$2$\mu$m$^2$ TMR element, of the structured sample.

Figure 6.8: I/V measurements of the 2$\times$2$\mu$m$^2$ TMR element.

[Original I/V curve.] [Differentiated I/V plot that is used for the BRINKMAN fit.]

As expected from chapter 5, the positioning system on top of the elements works quite well. Figure 6.9 shows two images of a video that was recorded during a positioning experiment. A single magnetic marker is drawn into the corner of the upper positioning structure by the magnetic gradient field. The images show the marker directly before (a) and after (b) the marker reaches the corner (center of the red circle marks the position of the bead). Although the conducting line is not embedded into the SiO$_2$, the positioning still works. Other recorded videos show that the conducting line is an obstacle, but many markers can easily leap over this barrier. It is no real drawback for the positioning system, therefore.

Figure 6.9: Placement of a single magnetic marker (MICROMOD marker with a diameter of 1.5$\mu$m) into the corner of the positioning structure, right before (a) and after (b) the marker reaches the final position. The images have a size of $86.7 \times 86.7$$\mu$m$^2$. See CD for the complete video.

[]     []

In the example of figure 6.9, the upper conducting line is not exactly aligned to the TMR elements, and so the marker is positioned beside the element and not directly on top of it (the TMR element in figure 6.9(b) is the square that is top left from the magnetic bead). In this situation, another effect helps to position the magnetic markers directly on top of the TMR elements. As can be seen in SEM images (confer figure 6.10), the stray field of the TMR element itself draws the magnetic particle on top of the element. So, in order to position a magnetic marker on a magnetoresistive element, the positioning system does not have to be very exact, because the stray field of the element does the rest.

Figure 6.10: SEM images of well positioned single magnetic markers (MICROMOD, $\varnothing$=1.5$\mu$m).

[Single marker on a 2$\times$2$\mu$m$^2$ TMR element, besides a big agglutination of markers.] [Single marker on a 4$\times$4$\mu$m$^2$ TMR element.]    

Figure 6.10 presents two SEM images as examples for a successful positioning. Figure 6.10(a) shows a single magnetic particle on a 2$\times$2$\mu$m$^2$ TMR element besides a big agglutination of markers on top of the corner of the positioning system. The second image (b) presents a single 1$\mu$m sized magnetic marker on a 4$\times$4$\mu$m$^2$ sized TMR element (it is the same element as in figure 6.9). This image reveals clearly the two local maxima on the sample. One bead is directly in the corner of the positioning system (lower right) and one bead is on the TMR element (center). Therefore, the stray field of the TMR element adds a local maximum to the magnetic gradient field of the positioning system.

Until today there is one major drawback of these experiments. No TMR element survived the drop of water during the positioning. Although the SiO$_2$ protection layer is made precisely, the water always seems to find a way to creep below it. On some samples, the destruction can be seen easily, but on others, the tunnel barrier just seems to be destroyed without reason. Figure 6.11 presents two examples for obvious destructions. The first image (a) shows some holes that originate from the positioning experiments. Right at the end of the experiments, when the water drop dried out, several holes all over the sample sprung up and short-circuited the top and bottom contacts of the TMR elements through the protection layer. The reason for this is yet unclear. The second image (b) shows some dark structures directly on top of the bottom contact line. It looks similar to the splintered glass that was presented in section 3.2. The water somehow finds a way below the SiO$_2$ protection layer and destroys the TMR elements. These were just two examples where the reason for the short-circuited TMR elements was obvious. Other samples showed no visible signs of destruction, but all TMR elements were always short-circuited after the positioning.

Figure 6.11: Destruction of the TMR elements.

[SEM image of holes that destroyed the sample.]     [Video image of water, destroying the bottom contact line of the TMR elements.]